OR18-03 Peripheral But Not Hepatic Insulin Resistance Associates With Adverse Vascular And Metabolic Health In Type 1 Diabetes: A Possible Target For Adjunctive Therapy

Abstract Disclosure: J.R. Snaith: None. N. Olsen: None. G. Kowalski: None. C. Bruce: None. J. Holmes-Walker: None. J.R. Greenfield: None. Insulin resistance in type 1 diabetes (T1D) is a cardiometabolic risk factor that is potentially targetable with adjunctive therapy. Prior studies of metformin in T1D have yielded mixed results. Whether liver and/or muscle insulin resistance is a prerequisite for metformin’s action in adults with T1D is unknown. The INTIMET study recruited 40 adults with T1D (age 37.4±8.8 years, HbA1c 7.5±0.9%, diabetes duration 22.9±8.8 years, BMI 26.4±0.9 kg/m2) and 20 age- and BMI-matched controls without diabetes. Subjects with T1D were randomized to metformin extended-release 1500mg daily or placebo for 26 weeks. Two-step hyperinsulinemic euglycemic clamps (low-dose 20 mU/m2/min, high-dose 60 mU/m2/min insulin) using 6,6-2H2-glucose tracers assessed liver (basal glucose rate of appearance [B-Ra] and low-dose endogenous glucose production [L-EGP]; primary outcomes) and muscle (high-dose glucose infusion rate [GIR]; secondary outcome) insulin resistance at t=0 in all subjects and at t=26 weeks in T1D. Dual-energy x-ray absorptiometry was used to measure fat free mass and visceral adipose tissue (VAT). Applanation tonometry was used to measure arterial stiffness (AIx, augmentation index). Adults with T1D and controls had similar B-Ra (18.2±5.1 vs 18.3±2.2 mmol/kgFFM/min; p=0.87). However, T1D were insulin-resistant at the liver (L-EGP 5.9±2.2 vs 3.6±1.7 mmol/kgFFM/min; p<0.001) and at muscle (GIR 62±20 vs 88±18 µmol/min·kgFFM; p<0.001). Despite these differences, VAT was comparable. In all subjects, baseline GIR correlated with L-EGP (r=-0.343; p=0.008), VAT (r=-0.269; p=0.039) and BMI (r=-0.304; p=0.019). In contrast, baseline L-EGP did not relate to VAT (r=0.060; p=0.653) or BMI (r=0.173; p=0.189). Adults with T1D had borderline higher AIx than controls (11.2±11.4 vs 4.7±13.1%, p=0.05). In the whole group, GIR was associated with LogAIx (r=-0.258, p=0.048), but L-EGP was not (r=0.154; p=0.246). In T1D, HbA1c correlated with GIR (r=-0.329; p=0.041) but L-EGP did not (r=0.307; p=0.057). After 26 weeks, there was no difference between metformin vs placebo-treated T1D subjects in HbA1c and weight, or B-Ra, L-EGP or GIR using generalised linear regression adjusted for age and baseline B-Ra, L-EGP and GIR. In these models, baseline GIR inversely predicted L-EGP (p=0.027) and directly predicted GIR (p<0.001) at 26 weeks. Adults with T1D exhibited greater liver and muscle insulin resistance, and higher arterial stiffness, which was not improved with metformin. Muscle but not liver insulin resistance related to BMI, VAT, HbA1c and AIx. Baseline muscle insulin resistance predicted muscle and hepatic insulin resistance at 26 weeks, irrespective of metformin treatment. This study highlights 1) the importance of muscle but not liver insulin resistance in T1D and 2) the need to identify muscle insulin resistant T1D individuals to enable interventions that improve vascular and metabolic outcomes. Presentation: Saturday, June 17, 2023